Emergent Misalignment As Prompt Sensitivity: a Research Note

Betley et al. 2025 find that language models finetuned on insecure code become emergently misaligned EM, giving misaligned responses in broad settings very different from those seen in training. However, it remains unclear as to why emergent misalignment occurs. We evaluate insecure models across...

SUSE CVE-2025-3611

Mattermost versions 10.7.x = 10.7.0, 10.5.x = 10.5.3, 9.11.x = 9.11.12 fails to properly enforce access control restrictions for System Manager roles, allowing authenticated users with System Manager privileges to view team details they should not have access to via direct API requests to team...

When There Is No Decoder: Removing Watermarks from Stable Diffusion Models in a No-Box Setting

Watermarking has emerged as a promising solution to counter harmful or deceptive AI-generated content by embedding hidden identifiers that trace content origins. However, the robustness of current watermarking techniques is still largely unexplored, raising critical questions about their...

SUSE CVE-2025-38099

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: Disable SCO support if READVOICESETTING is unsupported/broken A SCO connection without the proper voicesetting can cause the controller to lock up...

DEBIAN-CVE-2025-38099

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: Disable SCO support if READVOICESETTING is unsupported/broken A SCO connection without the proper voicesetting can cause the controller to lock up...

UBUNTU-CVE-2025-38099

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: Disable SCO support if READVOICESETTING is unsupported/broken A SCO connection without the proper voicesetting can cause the controller to lock up...

CVE-2025-38099 Bluetooth: Disable SCO support if READ_VOICE_SETTING is unsupported/broken

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: Disable SCO support if READVOICESETTING is unsupported/broken A SCO connection without the proper voicesetting can cause the controller to lock up...

Linux kernel 安全漏洞

Linux kernel is the kernel used by Linux, the open source operating system of the Linux Foundation in the United States. A security vulnerability exists in the Linux kernel that stems from Bluetooth SCO support not being disabled when READVOICESETTING is not supported, which could lead to...

jetty-http2-common: Jetty HTTP/2 Header List Size Vulnerability

A flaw was found in Eclipse Jetty. This vulnerability allows denial of service attack via an HTTP/2 client specifying a very large value for the SETTINGSMAXHEADERLISTSIZE parameter...

Pilz IndustrialPI 代码问题漏洞

Pilz IndustrialPI is a gateway for the Industrial Internet of Things from Pilz Individual Developers in Germany. A code issue vulnerability exists in Pilz IndustrialPI that stems from an unauthenticated login bypass resulting in a setting change...

MikroTik RouterOS 安全漏洞

MikroTik RouterOS is a Linux-based router operating system developed by the Latvian company MikroTik. The system can be deployed in a PC to make it provide router functionality. A security vulnerability exists in MikroTik RouterOS 7 versions prior to 7.14, which stems from a default setting error...

CVE-2025-52992

The Nix, Lix, and Guix package managers fail to properly set permissions when a derivation build fails. This may allow arbitrary processes to modify the content of a store outside of the build sandbox. This affects Nix before 2.24.15, 2.26.4, 2.28.4, and 2.29.1; Lix before 2.91.2, 2.92.2, and...

Tenda AC6 安全漏洞

The Tenda AC6 is a wireless router from the Chinese company Tenda. A buffer overflow vulnerability exists in Tenda AC6 V15.03.05.19 and earlier versions, which stems from the fromNatStaticSetting function failing to properly validate the length of the input data, and can be exploited by remote...

CVE-2025-20264

A vulnerability in the web-based management interface of Cisco Identity Services Engine ISE could allow an authenticated, remote attacker to bypass the authorization mechanisms for specific administrative functions. This vulnerability is due to insufficient authorization enforcement mechanisms fo...

Anti-Phishing Training Does Not Work: a Large-Scale Empirical Assessment of Multi-Modal Training Grounded in the NIST Phish Scale

Social engineering attacks using email, commonly known as phishing, are a critical cybersecurity threat. Phishing attacks often lead to operational incidents and data breaches. As a result, many organizations allocate a substantial portion of their cybersecurity budgets to phishing awareness...

Unlearning-Enhanced Website Fingerprinting Attack: against Backdoor Poisoning in Anonymous Networks

Website Fingerprinting WF is an effective tool for regulating and governing the dark web. However, its performance can be significantly degraded by backdoor poisoning attacks in practical deployments. This paper aims to address the problem of hidden backdoor poisoning attacks faced by Website...

Enhancing One-run Privacy Auditing with Quantile Regression-Based Membership Inference

Differential privacy DP auditing aims to provide empirical lower bounds on the privacy guarantees of DP mechanisms like DP-SGD. While some existing techniques require many training runs that are prohibitively costly, recent work introduces one-run auditing approaches that effectively audit DP-SGD...

CVE-2025-6019

A Local Privilege Escalation LPE vulnerability was found in libblockdev. Generally, the "allowactive" setting in Polkit permits a physically present user to take certain actions based on the session type. Due to the way libblockdev interacts with the udisks daemon, an "allowactive" user on a syst...

CVE-2025-6019

CVE-2025-6019 is a local privilege escalation in libblockdev that leverages the interaction with the udisks daemon and the Polkit “allow_active” setting to allow a physically present user to escalate to root. The issue arises when an attacker crafts an XFS image containing a SUID-root shell and m...

SUSE CVE-2022-50173

In the Linux kernel, the following vulnerability has been resolved: drm/msm/mdp5: Fix global state lock backoff We need to grab the lock after the early return for !hwpipe case. Otherwise, we could have hit contention yet still returned 0. Fixes an issue that the new CONFIGDRMDEBUGMODESETLOCK stu...